Machine Learning Crash Course | Google Developers
WHAT IS TENSORFLOW?
TensorFlow is an open source machine learning platform developed by Google. It has a JavaScript library for deploying machine learning models in the browser, a lite version for mobile and IoT devices, and an extended version for large production environments.TensorFlow is a framework for building, training, and deploying machine learning models. Kerasis a wrapper built on top of TensorFlow making it a bit more accessible, easier, and cleaner to work with.
WHAT IT DOES ?
1.Tensorflow provides several add-on libraries and resources to deploy your production models anywhere. 2.Through multiple levels of abstraction, tensorflowgives you an easy and flexible model building experience suitable for both experts and beginners.3.By using high-performance APIs, you have full control of your advanced models to quickly debug or effortlessly prototype these models.
TENSOR FLOW POSSESSED TOOLS
TensorFlow possessed multiples tools to produce machine learning systems:
TensorFlow(dah)
TensorFlow Core | Machine Learning for Beginners and Experts
•TensorFlow.js
TensorFlow.js | Machine Learning for Javascript Developers
•TensorFlow lite to deploy machine learning model in an embedded system
TensorFlow Lite | ML for Mobile and Edge Devices
•TensorFlow Extendedto productizing machine learning pipeline
TensorFlow Extended (TFX) | ML Production Pipelines
•TensorFlow Quantuma “library for rapid prototyping of hybrid quantum-classical ML models”
TensorFlow, how to build a ML model?
To interact with TensorFlow, one of the most popular API is the Python one (and to be honest that’s the one that I am the more comfortable with), but there aretwo paths availableto interact with this API:•The beginner one that is using a user-friendly sequential API called Keras•The expert one that is using a subclassing API more pythonicKerasis an API that can run on top of various ML frameworks as TensorFlow, CNTK and Theano to help people to easily reused functions to build layer, solver etcwithout going too deep on the ml framework